CN114430490A - Live question and answer and interface display method and computer storage medium - Google Patents

Live question and answer and interface display method and computer storage medium Download PDF

Info

Publication number
CN114430490A
CN114430490A CN202210067927.1A CN202210067927A CN114430490A CN 114430490 A CN114430490 A CN 114430490A CN 202210067927 A CN202210067927 A CN 202210067927A CN 114430490 A CN114430490 A CN 114430490A
Authority
CN
China
Prior art keywords
answer
question
live broadcast
live
audience
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210067927.1A
Other languages
Chinese (zh)
Inventor
林旭鸣
蒲黎明
崔雨豪
赵中州
周伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba China Co Ltd
Original Assignee
Alibaba China Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba China Co Ltd filed Critical Alibaba China Co Ltd
Priority to CN202210067927.1A priority Critical patent/CN114430490A/en
Publication of CN114430490A publication Critical patent/CN114430490A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/21—Server components or server architectures
    • H04N21/218—Source of audio or video content, e.g. local disk arrays
    • H04N21/2187—Live feed
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
    • H04N21/4312—Generation of visual interfaces for content selection or interaction; Content or additional data rendering involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/435—Processing of additional data, e.g. decrypting of additional data, reconstructing software from modules extracted from the transport stream
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/454—Content or additional data filtering, e.g. blocking advertisements
    • H04N21/4545—Input to filtering algorithms, e.g. filtering a region of the image
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/475—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data
    • H04N21/4758—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data for providing answers, e.g. voting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application
    • H04N21/4788—Supplemental services, e.g. displaying phone caller identification, shopping application communicating with other users, e.g. chatting

Abstract

The embodiment of the application provides a live question and answer method, a live interface display method and a computer storage medium, wherein the live question and answer method is suitable for a live room for live broadcast by using a virtual anchor, and the method comprises the following steps: acquiring a current live broadcast object of virtual main broadcast explanation in a live broadcast room and user information of a plurality of audience users, wherein the user information comprises attribute information and/or user preference information of the audience users; determining different recommended questions for the plurality of audience users according to the current live broadcast object, the user information and a question-answer mapping relation, wherein the question-answer mapping relation represents the corresponding relation between a plurality of live broadcast objects and a plurality of question-answer pairs; displaying different recommendation problems on live broadcast interfaces of different audience users to recommend the different recommendation problems to the corresponding audience users; and displaying the corresponding recommended answer on the live broadcast interface of the audience user according to the feedback input of the audience user.

Description

Live question and answer and interface display method and computer storage medium
Technical Field
The embodiment of the application relates to the technical field of computers, in particular to a live question answering method for a virtual anchor, a live interface display method and a computer storage medium.
Background
With the wide application of audio and video technologies, more and more services are realized by means of live broadcasting, such as commodity selling, content popularization and the like. At present, most of live broadcasts are played by real persons as anchor broadcasts, and the live broadcasts can explain live targets (such as commodities or other contents) and interact with audiences and answer audience questions. When a real person acts as a director, the director can professionally and deeply explain and explain the audience consultation problem in the live broadcast room, and sometimes can display the problems through pictures, such as displaying the colors of commodities or demonstrating the using mode.
But the live person cannot achieve a long uninterrupted anchor and some live rooms may also not be able to please the live anchor host. Thus, the virtual anchor takes place at the discretion. The virtual anchor is usually live based on a pre-constructed script, and the live broadcast is influenced by a content source of the pre-constructed script, so that the virtual anchor is limited in dealing with and handling problems of audiences in a live broadcast room.
Disclosure of Invention
In view of the above, embodiments of the present application provide a live question and answer and live room interface display scheme to at least partially solve the above problems.
According to a first aspect of an embodiment of the present application, a live question and answer method is provided, which is applicable to a live broadcast room using a virtual anchor for live broadcast, and the method includes: acquiring a current live broadcast object of virtual main broadcast explanation in a live broadcast room and user information of a plurality of audience users, wherein the user information comprises attribute information and/or user preference information of the audience users; determining different recommended questions for the plurality of audience users according to the current live broadcast object, the user information and a question-answer mapping relation, wherein the question-answer mapping relation represents the corresponding relation between a plurality of live broadcast objects and a plurality of question-answer pairs; displaying different recommendation problems on live broadcast interfaces of different audience users to recommend the different recommendation problems to the corresponding audience users; and displaying the corresponding recommended answer on the live broadcast interface of the audience user according to the feedback input of the audience user.
According to a second aspect of the embodiments of the present application, a live broadcast room interface display method is provided, which is applicable to a live broadcast room in which a virtual anchor is used for live broadcast, and the method includes: after detecting that a plurality of audience users enter a live broadcast room, displaying a live broadcast room interface to the plurality of audience users; and displaying a live broadcast object currently explained by a virtual anchor in the live broadcast room in a first area of a live broadcast room interface of different audience users, and displaying different recommendation problems aiming at the live broadcast object in a second area of the live broadcast room interface aiming at different audience users.
According to a third aspect of embodiments of the present application, there is provided an electronic apparatus, including: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface are communicated with each other through the communication bus; the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the corresponding operation of the method according to the first aspect or the second aspect.
According to a fourth aspect of embodiments of the present application, there is provided a computer storage medium having stored thereon a computer program which, when executed by a processor, implements a method according to the first or second aspect.
According to the scheme provided by the embodiment of the application, aiming at a live broadcast room for live broadcast by using a virtual anchor, different recommendation problems are determined for a plurality of audience users on the basis of live broadcast objects currently explained by the virtual anchor, user information of the audience users and question-answer mapping relations during live broadcast. And then, feeding back corresponding recommended answers to the audience users according to feedback input of different audience users. Therefore, according to the scheme of the embodiment of the application, firstly, for audience users entering a live broadcast room, recommendation questions can be automatically displayed on the live broadcast room interface, so that the cost of asking questions by the audience users can be reduced, and the efficiency of asking questions by the audience users is improved; secondly, various factors such as currently explained live broadcast objects, user information of each audience user, question-answer mapping relations and the like are comprehensively considered for the recommendation problems displayed by different audience users, so that the recommendation problems are more in line with the possible requirements of the audience users, and the live broadcast watching experience of the audience users in a live broadcast room is enhanced; moreover, compared with the problem that the audience user inputs the question by himself, the scheme of the application can combine the content currently explained by the virtual anchor, actively guide and prompt the audience user to ask the question by recommending the question, simplify the question asking process, and improve the interaction efficiency of the audience user and the virtual anchor. Therefore, the virtual anchor can more flexibly deal with and process the problems of the audience of the live broadcast room.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present application, and other drawings can be obtained by those skilled in the art according to the drawings.
Fig. 1 is a schematic diagram of an exemplary system to which a live question answering method according to an embodiment of the present application is applied;
fig. 2A is a flowchart illustrating steps of a live question answering method according to an embodiment of the present application;
FIG. 2B is a diagram of a question-answer mapping relationship in the embodiment shown in FIG. 2A;
FIG. 2C is a diagram illustrating an example of a scenario in the embodiment shown in FIG. 2A;
fig. 3A is a flowchart illustrating steps of a live question answering method according to a second embodiment of the present application;
FIG. 3B is a diagram illustrating a process of establishing a mapping relationship in the embodiment shown in FIG. 3A;
FIG. 3C is a schematic diagram illustrating a video or image matching process in a live question answering process in the embodiment shown in FIG. 3A;
FIG. 3D is a diagram illustrating a process of live question answering in the embodiment shown in FIG. 3A;
fig. 4A is a flowchart illustrating steps of a live broadcast interface displaying method according to a third embodiment of the present application;
FIG. 4B is a schematic diagram of an example of an interface in the embodiment of FIG. 4A;
fig. 5 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present application.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application shall fall within the scope of the protection of the embodiments in the present application.
The following further describes specific implementations of embodiments of the present application with reference to the drawings of the embodiments of the present application.
Fig. 1 illustrates an exemplary system to which a live question answering method according to an embodiment of the present application is applied. As shown in fig. 1, the system 100 may include a server 102, a communication network 104, and/or one or more viewer-side user devices 106, illustrated in fig. 1 as a plurality of viewer-side user devices.
Server 102 may be any suitable server for storing information, data, programs, and/or any other suitable type of content. In some embodiments, server 102 may perform any suitable functions. For example, in some embodiments, the server 102 may be configured to determine different recommendation questions for different viewer users and make recommendations to the corresponding viewer users. As an alternative example, in some embodiments, the server 102 may determine different recommendation questions for multiple viewer users according to a current live broadcast object explained by the virtual anchor, user information of the multiple viewer users, and a question-answer mapping relationship in a live broadcast room in which the virtual anchor is live. As another example, in some embodiments, the server 102 may be configured to determine a corresponding recommendation answer according to a feedback input of the audience user to a recommendation question presented on a respective live-room interface, and present the recommendation answer through the live-room interface of the audience user.
In some embodiments, the communication network 104 may be any suitable combination of one or more wired and/or wireless networks. For example, the communication network 104 can include any one or more of the following: the network may include, but is not limited to, the internet, an intranet, a Wide Area Network (WAN), a Local Area Network (LAN), a wireless network, a Digital Subscriber Line (DSL) network, a frame relay network, an Asynchronous Transfer Mode (ATM) network, a Virtual Private Network (VPN), and/or any other suitable communication network. The user device 106 can be connected to the communication network 104 by one or more communication links (e.g., communication link 112), and the communication network 104 can be linked to the server 102 via one or more communication links (e.g., communication link 114). The communication link may be any communication link suitable for communicating data between the user device 106 and the server 102, such as a network link, a dial-up link, a wireless link, a hardwired link, any other suitable communication link, or any suitable combination of such links.
The audience member user devices 106 may include any user device or devices adapted to present an interface and interact with an audience through the interface. In some embodiments, the viewer-side user device 106 may present a corresponding live-air interface upon detecting that multiple viewer users are in a live-air room; and displaying the live broadcasting object currently explained by the virtual main broadcasting in a first area of a live broadcasting room interface, and displaying different recommendation problems aiming at different audience users in a second area of the live broadcasting room interface. In some embodiments, the viewer-side user device 106 may further receive feedback input of the viewer user for the recommendation question, and send information of the feedback input to the server 102, so as to obtain a corresponding recommendation answer from the server 102, and then explain the recommendation answer through a virtual anchor, and/or display the recommendation answer through a live program interface. In some embodiments, the viewer-side user device 106 may comprise any suitable type of device. For example, in some embodiments, the spectator user devices 106 may include mobile devices, tablet computers, laptop computers, desktop computers, wearable computers, game consoles, media players, vehicle entertainment systems, and/or any other suitable type of user device.
Although server 102 is illustrated as one device, in some embodiments, any suitable number of devices may be used to perform the functions performed by server 102. For example, in some embodiments, multiple devices may be used to implement the functions performed by the server 102. Alternatively, the functionality of the server 102 may be implemented using a cloud service.
Based on the system, the embodiment of the application provides a live question-answering scheme and a live session interface scheme, and the following description is given through a plurality of embodiments.
Example one
Referring to fig. 2A, a flowchart illustrating steps of a live question answering method according to an embodiment of the present application is shown.
The live broadcast question-answering method of the embodiment is explained from a server side, and the live broadcast question-answering method is suitable for a live broadcast room for live broadcast by using a virtual anchor, and is suitable for various types of live broadcasts, such as e-commerce, education, culture, sports, news, knowledge, life, electronic contests and the like. The method comprises the following steps:
step S202: the method includes the steps of obtaining a current live broadcast object of virtual main broadcast explanation in a live broadcast room and user information of a plurality of audience users.
Wherein the user information includes attribute information and/or user preference information of the viewer user, and the user information can be obtained under the condition of obtaining the information use permission. The attribute information of the viewer user is used to represent basic attributes of the viewer user, including but not limited to: identification of audience users, city, age, specialty, etc. The user preference information is used to indicate the preference of the user for live broadcasting, and includes but is not limited to: question preferences, live object preferences, etc. The user preference information can be obtained based on the analysis of historical question data of audience users, historical interaction data with the anchor, historical browsing behavior data, historical comment data and the like. Obviously, different viewer users have different user information.
In the embodiments of the present application, the numbers "plural" and "plural" relating to "plural" mean two or more unless otherwise specified.
Step S204: and determining different recommendation problems for the plurality of audience users according to the current live broadcast object, the user information of the plurality of audience users and the question-answer mapping relation. And the question-answer mapping relation represents the corresponding relation between a plurality of live broadcast objects and a plurality of question-answer pairs.
The server prestores corresponding relations between a plurality of live broadcast objects and a plurality of question and answer pairs, different live broadcast objects correspond to different question and answer pairs, and certainly, the condition that the question and answer pairs corresponding to different live broadcast objects are partially overlapped is possible to exist, which is also within the protection scope of the application. For each live object in the live objects, a mapping relation exists between the live object and the question-answer pairs. The mapping relationship may be pre-established based on historical live data.
Based on the above, after the live broadcast object currently being explained by the virtual main broadcast in the live broadcast room is obtained, a plurality of question and answer pairs corresponding to the live broadcast object can be determined based on the live broadcast object and the question and answer mapping relation. And determining question-answer pairs matched with the user information of each audience user for the audience users according to the user information of the audience users, and taking the questions in the matched question-answer pairs as the recommendation questions of the audience users. In this case, the specific implementation of matching question-answer pairs based on user information can be implemented by those skilled in the art in an appropriate manner according to the actual situation, including but not limited to: the method is realized by a machine learning model which is completed by training and takes user information and question-answer pairs as input and takes the matching degree of the user information and question-answer pairs as output, or by a similarity algorithm of information matching, and the like.
Exemplarily, it is assumed that the virtual host is currently explaining the live object X, and the server pre-stores the corresponding relationship between the live object X and the question- answer pairs 1, 2, 3, 4, and 5. It should be noted here that the live object currently explained by the virtual anchor may be completely the same as the live object in the pre-stored mapping relationship, but may also be a live object in the question-answer mapping relationship that is matched with the live object currently explained when the similarity between the live object and the pre-stored mapping relationship reaches a certain degree.
Further, assuming that audience users A, B and C correspond to respective user information a ', B ' and C ', respectively, the degree of matching of the respective user information with the question-answer pair is calculated as shown in table 1 below:
Figure BDA0003480899780000051
as can be seen, for the current live object X, the question-answer pair 3 is most matched with the audience user a, the question-answer pair 1 is most matched with the audience user B, and the question-answer pair 5 is most matched with the audience user C. Based on this, the questions in question-answer pair 3 are determined as recommended questions to be recommended to viewer user a, the questions in question-answer pair 1 are determined as recommended questions to be recommended to viewer user B, and the questions in question-answer pair 5 are determined as recommended questions to be recommended to viewer user C. It should be noted that, in this example, only one question-answer pair is taken as an example, but it should be understood by those skilled in the art that, in practical applications, a threshold of matching degree may be set, and all questions exceeding the threshold of matching degree may be taken as recommendation questions, that is, one or more recommendation questions for one audience user may be taken. The matching degree threshold may be flexibly set by a person skilled in the art according to actual requirements, and the embodiment of the present application does not limit this.
In one possible approach, a question-answer intention element is further added to the question-answer mapping relationship, that is, the question-answer mapping relationship may represent correspondence between a plurality of live broadcast objects and a plurality of question-answer pairs and a plurality of question-answer intentions. For example, live object X corresponds to question- answer pairs 1, 2, and 3, and question- answer pairs 1, 2, and 3 correspond to question-answer intentions 1X, 2X, and 3X, respectively, and then a question-answer mapping relationship as shown in fig. 2B is formed.
Based on this, in a preferred manner, the present step can be implemented as: for each viewer user: analyzing intentions of the audience users according to the current live broadcast objects and the user information of the audience users to obtain corresponding user intentions; obtaining a plurality of question-answer pairs corresponding to the current live broadcast object according to the question-answer mapping relation; acquiring a plurality of question-answer intention information corresponding to a plurality of question-answer pairs; and determining the recommendation questions corresponding to the audience users according to the matching degree of the user intentions and the question-answer intention information. Through the intention analysis mode, the user intention is matched with the question and answer intention information in the question and answer mapping relation, the implementation is simple, and the efficiency of determining question and answer pairs for audience users and then determining the recommended questions from the question and answer pairs can be greatly improved.
Step S206: and displaying different recommendation problems on the live broadcast interfaces of different audience users so as to recommend the different recommendation problems to the corresponding audience users.
After determining different recommendation problems for different audience users, the server can send information of each audience user and corresponding recommendation problems to the live broadcast platform, so as to send the information to the audience client of each audience user through the live broadcast platform, and display the information through the live broadcast interface of each audience user. Or, the server can also recommend the question to send the live broadcasting interface of the corresponding audience user for displaying. Because different audience users have different corresponding recommendation problems, the display of the live broadcast room interfaces of different audience users is different.
Step S208: and displaying the corresponding recommended answer on the live broadcast interface of the audience user according to the feedback input of the audience user.
For each audience user, the audience user can see the recommendation problem through the live broadcast room interface of the audience user, and further, if the audience user thinks that the recommendation problem reflects the problem that the audience user wants to consult, the audience user can perform corresponding feedback input operation, such as operation of clicking the recommendation problem, and the like. The information of the feedback input operation can be sent to a server, the server determines a corresponding recommended answer according to the feedback input, and then the recommended answer is directly sent to the live broadcast room of the audience user, or the recommended answer is sent to the live broadcast room of the audience user through a live broadcast platform so as to be displayed through the live broadcast room interface of the audience user.
The feedback input is not limited to the above operation of directly clicking on the recommendation question, and in practical applications, the feedback input may be implemented by a viewer user who rewrites the question based on the recommendation question. Similarly, the rewritten question is finally sent to a server, the server determines a corresponding recommended answer according to the rewritten question, and feeds the recommended answer back to the live broadcast room of the audience user in a live broadcast mode or feeds back to the live broadcast room of the audience user through a live broadcast platform.
After obtaining the recommendation answer, in one possible approach, the recommendation answer may be presented at the viewer-user's live room interface and/or driven to be spoken by the virtual main play. And the watching and acquisition of new live broadcast contents by audience users can not be influenced by adopting a live broadcast room interface display mode. By adopting the virtual main broadcasting explaining mode, the experience of audience users can be effectively improved. Combining the two can make the audience user more effectively understand the answers to the questions they consult.
However, in order to enable the question answers and answers to be displayed more intuitively and to be more convenient for the user to understand, in a feasible manner, a multi-mode display manner can be adopted, namely, a matched video or image is determined based on the recommended questions and the recommended answers; the video or image is presented at the audience user's live room interface.
And determining the matched video or image based on the semantic labels respectively corresponding to the recommended question and the recommended answer. And when the matching is specifically carried out, the matching can be realized by adopting a trained machine learning model. The semantic labels can effectively reflect the semantics of the recommended questions and the recommended answers, and for videos or images, the videos or images also have certain meanings or expression themes, so that the videos or images matched with the question-answer pairs of the recommended questions and the recommended answers can be determined according to the matching degree between the semantics and the meanings or expression themes of the images, and information enrichment and supplement are achieved.
The above process is exemplified below in a specific scenario, as shown in fig. 2C.
In fig. 2C, it is assumed that the virtual anchor is currently explaining a down jacket Y, and the server pre-stores the corresponding relationship between the down jacket Y and the question- answer pairs 1, 2, 3, and 4. Wherein, the question-answer pair 1 is a question-answer pair of the down content question and the answer of the down jacket Y, the question-answer pair 2 is a question-answer pair of the color question and the answer of the down jacket Y, the question-answer pair 3 is a question-answer pair of the type question and the answer of the down jacket Y, the question-answer pair 4 is a question-answer pair of the material question and the answer of the down jacket Y, and the question-answer pair 5 is a question-answer pair of the cleaning question and the answer of the down jacket Y.
Further, assuming that the viewer users A, B and C correspond to the respective user information a ', B ', and C ', the results shown in table 1 are obtained by calculating the matching degree between the respective user information and the question-answer pair.
Then, based on the results, it can be determined that for the current live object X, the question-answer pair 3 is the closest match to viewer user a, the question-answer pair 1 is the closest match to viewer user B, and the question-answer pair 5 is the closest match to viewer user C.
Based on this, questions and answers can be answered to questions in 3, such as "how does the down jacket fit? "determine as a recommended question to recommend to audience user a, question and answer the question in pair 1, e.g.," what the down jacket is down? "determine as a recommended question to recommend to viewer user B, question and answer the question in 5" how to wash this down jacket? "determined as a recommendation question to make a recommendation to viewer user C.
Further, audience users A, B and C will each present their respective recommendation questions via their respective live room interfaces. Suppose again that if viewer user B clicks on the recommendation question "what down content is for this down jacket" presented on the live broadcast room interface? "the server determines the recommended answer corresponding to the recommended question based on question and answer pair 1, for example," the down jacket has a down content determined according to the model of the jacket, and there are three models of 160, 165 and 170, and the down content is 120 g, 140 g and 170 g, respectively ". This recommendation answer is eventually sent to viewer user B's live room, in this example, the recommendation answer is shown through the live room interface. However, as described above, the virtual host may be driven to explain the recommended answer, or both the recommended answer and the virtual host may be presented and the recommended answer may be explained.
According to the embodiment, aiming at the live broadcast room for live broadcast by using the virtual anchor, different recommendation problems are determined for a plurality of audience users on the basis of the live broadcast object currently explained by the virtual anchor, the user information of the audience users and the question-answer mapping relation during live broadcast. And then, feeding back corresponding recommended answers to the audience users according to feedback input of different audience users. Therefore, according to the scheme of the embodiment of the application, firstly, for audience users entering the live broadcast room, the recommendation questions can be automatically displayed on the live broadcast room interface, so that the cost of asking questions of the audience users can be reduced, and the efficiency of asking questions of the audience users is improved. Secondly, various factors such as currently explained live broadcast objects, user information of each audience user, question-answer mapping relations and the like are comprehensively considered for the recommendation problems displayed by different audience users, so that the recommendation problems are more in line with the possible requirements of the audience users, and the live broadcast watching experience of the audience users in a live broadcast room is enhanced; moreover, compared with the problem that the audience user inputs the question by himself, the scheme of the application can combine the content currently explained by the virtual anchor, actively guide and prompt the audience user to ask the question by recommending the question, simplify the question asking process, and improve the interaction efficiency of the audience user and the virtual anchor. Therefore, the virtual anchor can more flexibly deal with and process the problems of the audience of the live broadcast room.
Example two
Referring to fig. 3A, a flowchart illustrating steps of a live question answering method according to a second embodiment of the present application is shown.
The embodiment of the present invention explains the live question-answer method according to the embodiment of the present invention with emphasis on the construction of the question-answer mapping relationship. The live broadcast question answering method comprises the following steps:
step S302: and constructing a question-answer mapping relation.
As described above, in one mode, the question-answer mapping relationship is used to reflect the relationship between the live object and the question-answer pair, and for convenience of description, in this embodiment, the live object is represented as Item, and the question-answer pair is represented as (Q, a), where Q represents question data and a represents answer data. Based on this, the above question-answer mapping relationship can be expressed as { Item, (Q, a) }, where there are a plurality of (Q, a) items corresponding to one Item. In a feasible manner, the question-answer mapping relationship may be established by determining a live object (Item) corresponding to each question-answer pair (Q, a) based on question data Q and answer data a of a plurality of question-answer pairs.
Illustratively, in one mode, historical live broadcast data can be acquired, wherein the historical live broadcast data at least comprises live broadcast audio data and text data corresponding to the live broadcast audio data; performing question and answer recognition on the text data to obtain question and answer pairs in the text data; analyzing the question-answer pair to acquire the information of the corresponding live broadcast object; and establishing a mapping relation between the live broadcast object and the question-answer pair. The historical live broadcast data is generally live broadcast historical data of a real person, an answer mode surrounding a certain question can be identified, learned and mined from daily live broadcast data of a real person anchor through a live broadcast learning mode based on the real person, and due to the fact that the answer mode has rich professional information, the content richness is higher in content depth and professionality compared with the content richness obtained through other modes such as micro-panning articles, detail pages and the like, and the specialty is higher than that of a mode constructed through a knowledge map. Therefore, the constructed question-answer mapping relation has good professionality and content depth.
In a feasible mode, when the question-answer pairs are analyzed to obtain the information of the corresponding live broadcast object, the object identifiers in the question-answer pairs can be obtained through the analysis of the question-answer pairs; and obtaining the information of the live objects corresponding to the object identifiers according to the preset corresponding relation between the object identifiers and the live objects. By the method, the live broadcast object corresponding to the question-answer pair can be determined quickly and efficiently. Taking e-commerce scenes as an example, the object identifier can be a commodity name, a commodity link number, a commodity number and the like; taking the teaching scene as an example, the object identifier may be a knowledge point name, a knowledge point number, or the like.
However, in some cases, text data corresponding to live audio data may include some invalid information, and in order to improve the efficiency of constructing the question-answer mapping relationship, in a feasible manner, question-answer recognition is performed on the text data, and obtaining question-answer pairs in the text data may be implemented as: performing question-answer recognition on the text data to acquire all candidate question-answer pairs in the text data; performing semantic recognition on all candidate question-answer pairs; and screening effective question-answer pairs from all candidate question-answer pairs according to the result of semantic recognition. For example, all candidate question-answer pairs are screened, and question-answer pairs with strong effectiveness are filtered, such as real-time information question-answer pairs or preferential information question-answer pairs, and the screened question-answer pairs are used as effective question-answer pairs.
Although a more effective question-answer mapping relation can be constructed based on the effective question-answer pairs, in order to further improve the characteristic, pertinence and normalization of the question-answer pairs in the constructed question-answer mapping relation, question data and answer data in the question-answer pairs can be further processed. In one possible approach, valid answer data may be obtained from valid question-answer pairs; regenerating updated question data matched with the effective answer data based on the effective answer data; and taking the updated question data as question data of question-answer pairs. When the live broadcast is carried out by a real person, the anchor broadcast usually repeats questions asked by the user, so that the question and answer quality is poor, and therefore, only answer data are reserved through the method, and question reconstruction is carried out based on the answer data, so that reconstructed questions are more accurate and have higher quality. The regeneration and updating of the question data can be realized by adopting a machine learning model, namely a trained model which can generate the question data based on the answer data, such as an confrontation network model; the original question data may also be reprocessed based on the answer data, such as deduplication or semantic extraction, and updated question data may be generated based on the processing results.
In addition, in another possible manner, the association degree of the valid answer data in the valid question-answer pair with at least one key sentence may be determined; screening key answer short sentences according to the relevance; rewriting effective answer data in the effective question-answer pair according to the key answer short sentence to generate updated answer data, wherein the updated answer data comprises the key answer short sentence; and using the updated answer data as answer data of the question-answer pair. The key sentences may be phrases generated in advance according to information of historical live objects, different live objects may correspond to different phrases, and intersections may exist between the phrases of different live objects. In this way, the regenerated update answer data can be made more characteristic. In specific implementation, one or more sentences can be extracted from the effective answer data, then the relevance degree of the extracted sentences and the key sentences corresponding to the live broadcast object is calculated, the key answer short sentences are screened out according to the calculation result, and the effective answer data is rewritten based on the calculation result to generate updated answer data.
It should be noted that, in the above manner, the valid answer data means answer data in a valid question-answer pair.
In another mode of the present application, the question-answer mapping relationship is used for reflecting the relationship between the live object and the question-answer pair and the question-answer Intention, and may be represented as { Item, (Q, a), meaning }, where meaning represents the question-answer Intention. There are multiple (Q, A) for one Item, and one Intention for each (Q, A). In one possible way, the intetion may determine the question-answer Intention information corresponding to each question-answer pair based on the question data and answer data in a plurality of question-answer pairs. Representing the question-answer mapping relationship in the form of { Item, (Q, a), intent }, may facilitate subsequent determination of recommended questions for the viewer user.
It should be noted that the above-mentioned various modes can be used alternatively or in combination to achieve better effect.
For ease of understanding, the above process is illustrated below in a specific example of an e-commerce application, as shown in fig. 3B. The process comprises the following steps:
(1) and acquiring the live broadcast ASR corpus.
In live ASR text, a line represents a sentence. And (4) finding an anchor point by identifying a question, and using partial content behind the question as an answer. Assuming that the k-th sentence is recognized as a question sentence, the contents from the k + 1-th sentence to the k + n-th sentence (n > ═ 1) are taken as answers. For example, if n is 10, the first sentence is a question, and the subsequent 10 sentences are combined to form an answer, thereby forming one (Q, a) sample.
Furthermore, the link number corresponding to the (Q, a) sample is identified, and the link number is mapped to a related live broadcast object, such as a related commodity, and is used as the live broadcast object corresponding to the (Q, a) sample. For example, if "the toner linked to the x number of us has a moisturizing effect" in the sample, the product ID corresponding to each link number can be known through the mapping relationship between the product and the link number, and the product corresponding to the x number link can be found through the mapping relationship.
(2) Sentence level intent recognition is performed on the (Q, a) samples.
After the (Q, A) sample is obtained, QA fragments (such as data related to preferential activities, real-time information and the like) which are not suitable for retaining or cannot identify the intention are filtered through a semantic intention identification model, and an effective (Q, A) sample is screened out. The semantic intention model can be realized through a classification model, namely inputting (Q, A) and outputting a corresponding semantic intention label.
(3) Answer rewrite is performed based on the screened (Q, A) samples.
In the screened-out QA segment, there may be samples with poor quality of the problem Q (because anchor is a problem of repeated user questions in live webcast), so in this example, only the a segment is retained for the screened-out (Q, a) sample. Then, the question is regenerated by using a generative model according to the A segment, wherein the generative model inputs the answer A and outputs the question Q' aligned with the A.
(4) And (5) carrying out key phrase retrieval based on the screened (Q, A) sample.
As the original A segment still has a lot of contents irrelevant to the live broadcast object, sentence break and word mistake problems and the like, a matching model is introduced, the relevancy of each short sentence in the A and all core keywords of the live broadcast object is calculated, short sentences relevant to the live broadcast object are obtained through retrieval and sequencing, and A' is obtained through reservation.
(5) And (Q, A) samples subjected to key phrase retrieval are rewritten.
In this step, a and a 'are input into the NAR model, and the model has a function of obtaining a final rewritten candidate answer a ″ through multiple additions, deletions and alterations on the premise of retaining the content of a'.
(6) And constructing a question-answer mapping relation based on the (Q ', A') and the association relation between the (Q, A) and the live broadcast object obtained before.
If the question-answer mapping relationship further comprises a question-answer intention element, the question data and answer data can be obtained based on (Q ', A') or (Q, A); constructing a corresponding question and answer intention based on the question data and the answer data; and then inputting the question-answer intentions into an intention-question mapping module, and expanding the original live broadcast object-question-answer content into a live broadcast object-intention-question-answer content.
Step S304: the method includes the steps of obtaining a current live broadcast object of virtual main broadcast explanation in a live broadcast room and user information of a plurality of audience users.
Wherein the user information comprises attribute information and/or user preference information of the viewer user.
Step S306: and determining different recommendation questions for a plurality of audience users according to the current live broadcast object, the user information of the audience users and the question-answer mapping relation.
The question-answer mapping relationship may be the mapping relationship established in step S302.
If the question-answer mapping relationship also comprises a question-answer intention element, when a recommended question is determined, a plurality of corresponding live broadcast object-question-answer intention-question-answer pairs can be found in a live broadcast object-question-answer intention-question-answer content library through a live broadcast object, the questions which are possibly clicked by audience users are pushed through a personalized sorting model by combining user information of the audience users, and the audience users can directly send the questions in a comment area through clicking.
Step S308: and displaying different recommendation problems on the live broadcast interfaces of different audience users so as to recommend the different recommendation problems to the corresponding audience users.
The specific implementation of the above steps S304-S306 can refer to the description of the relevant parts in the foregoing embodiments, and will not be described herein again.
Step S310: and displaying the corresponding recommended answers on the interface of the live broadcasting room of the audience user according to the feedback input of the audience user.
After feedback input of the audience user is received, for example, the audience user clicks a recommendation question and sends the recommendation question in the comment area, a recommendation answer corresponding to the recommendation question can be matched.
Similar to the scheme in the first embodiment, in this embodiment, in addition to determining the recommendation question, a matching video or image may be determined based on the recommendation question and the recommendation answer, and the video or image may be displayed in the live-air interface of the audience user.
Illustratively, as shown in fig. 3C, the process of matching the corresponding video or image is as follows: firstly, acquiring visual materials (videos or images) of a live object, and screening the visual materials to filter out meaningless or unqualified materials; and acquiring text data corresponding to the live broadcast object, and performing paragraph splitting analysis on the text data to acquire each QA fragment and key information (such as keywords, words, short sentences and the like) of the QA fragment.
The following processes in this example are all exemplified by one QA fragment, and it will be understood by those skilled in the art that each QA fragment may use the same process to obtain corresponding visual material. The images in the visual materials can comprise main and auxiliary images of a live object, SKU images, detail page images and images in some grass articles; the video may include a main picture video of a live object, a seed grass video, a live room explanation video, and some image-generated video.
And secondly, accurately matching the image based on a text matching model, and matching the video material text with QA key information of a live broadcast object. Wherein the video material text may be from a main text on the video image. If the visual material cannot be matched, the following processing is performed.
Then, mapping and matching are carried out on the QA key information of the live broadcast object based on the category label of the visual material. For example, if the QA key information indicates that the QA fragment is material-speaking, the material class picture is displayed, and if the visual material cannot be matched, the following processing is performed.
And then, directly judging whether the visual material is matched with the QA key information or not based on the multi-mode matching map, and if the visual material cannot be matched, performing the following processing.
Finally, if none of the above matches match the visual material, a preset configuration diagram (a general display configuration diagram shown in the figure) is used, that is, an exquisite image or video with specific semantics is displayed by default.
Through the processing, a question-answer QA fragment is input, and each sentence in the QA fragment may have some semantic tags; the output is multimodal data with video or images fitted for each sentence. After reading the related visual materials of the live broadcast object, recalling images or matching videos for each QA fragment in the corresponding question and answer content.
By the mode, when the virtual anchor explains commodities in the live broadcast room, the visual materials related to the live broadcast object can be presented in the live broadcast room in an image or video mode in combination with the character script. Therefore, on one hand, the explanation effect and the live broadcast atmosphere of the live broadcast object of the virtual anchor are enhanced, and on the other hand, the visual feeling of the audience user on the live broadcast object in the live broadcast room can be improved.
In the following, a general process of the live question answering method provided by the embodiment of the present application is schematically described in combination with question-answer mapping relationship construction, multi-modal data presentation, and recommendation question determination, as shown in fig. 3D.
In fig. 3D, a question-answer mapping relationship is learned from live ASR of a real person, and a visual material library is created to present visual material presentation contents (such as images, videos, etc.) aligned with the question-answer semantics to audience users while performing real-time question-answer. Based on the constructed question-answer mapping relation, retrieving questions related to the current live broadcast object explained by the virtual main broadcast, determining recommended questions by combining user information of audience users, and recommending the questions to the audience users so as to actively inform and guide the questions which can be asked by the audience users through front-end display. After feedback of audience users for the recommendation questions is obtained, retrieval is carried out based on the question-answer mapping relation and the visual material library, and recommendation answers matched with the recommendation questions and corresponding visual materials are obtained. Finally, the answers corresponding to the recommended questions and the corresponding visual materials are displayed to the audience users in a one-to-one mode or a one-to-many mode (for example, the audience users directionally reply each question in a pop-up window mode, namely, the questions of a plurality of audience users can be answered at one time), and the audience users are asked and puzzled.
According to the embodiment, aiming at the live broadcast room for live broadcast by using the virtual anchor, different recommendation problems are determined for a plurality of audience users on the basis of the live broadcast object currently explained by the virtual anchor, the user information of the audience users and the question-answer mapping relation during live broadcast. And further, feeding back corresponding recommended answers to the audience users according to feedback input of different audience users. Therefore, according to the scheme of the embodiment of the application, firstly, for audience users entering a live broadcast room, recommendation questions can be automatically displayed on the live broadcast room interface, so that the cost of asking questions by the audience users can be reduced, and the efficiency of asking questions by the audience users is improved; secondly, various factors such as currently explained live broadcast objects, user information of each audience user, question-answer mapping relations and the like are comprehensively considered for the recommendation problems displayed by different audience users, so that the recommendation problems are more in line with the possible requirements of the audience users, and the live broadcast watching experience of the audience users in a live broadcast room is enhanced; moreover, compared with the problem that the audience user inputs the question by himself, the scheme of the application can combine the content currently explained by the virtual anchor, actively guide and prompt the audience user to ask the question by recommending the question, simplify the question asking process, and improve the interaction efficiency of the audience user and the virtual anchor. Therefore, the virtual anchor can more flexibly deal with and process the problems of the audience of the live broadcast room.
EXAMPLE III
Referring to fig. 4A, a flowchart illustrating steps of a live broadcast interface presentation method according to a third embodiment of the present application is shown.
The present embodiment explains the scheme of the embodiment of the present application from the perspective of presentation of a live broadcast interface. The method for displaying the interface of the live broadcast room is suitable for the live broadcast room which utilizes the virtual anchor to carry out live broadcast, and comprises the following steps:
step S402: and after detecting that the plurality of audience users enter the live broadcast room, displaying a live broadcast room interface for the plurality of audience users.
Step S404: and displaying the live broadcasting object currently explained by the virtual main broadcasting in the live broadcasting room in a first area of the live broadcasting room interface of different audience users, and displaying different recommendation problems aiming at the live broadcasting object aiming at different audience users in a second area of the live broadcasting room interface.
In this embodiment, because the recommendation problems corresponding to different viewer users are different, the display content of the second area in the live broadcast room interface is also different.
An exemplary live broadcast room interface is shown in fig. 4B, where the upper left side interface in fig. 4B is a live broadcast room interface of an audience user a, and the upper right side interface in fig. 4B is a live broadcast room interface of an audience user B, and thus, specific recommendation questions displayed in the second area in the live broadcast room interfaces of the audience users a and B are different. Assuming that the viewer users a and B both perform click operations based on the recommendation question, as indicated by the hand shape in the figure, the corresponding recommendation answers will also be displayed in the second area, as shown in the left interface of the lower half of the figure and the right interface of the lower half of the figure, respectively. In addition, the virtual anchor also explains their respective recommended answers to audience users a and B, respectively. That is, the method of this embodiment may further include: receiving feedback input of audience users for the displayed recommendation questions; acquiring a recommendation answer corresponding to the recommendation question according to the feedback input; and explaining the recommended answers through the virtual main broadcasting and/or displaying the recommended answers through a live broadcasting room interface.
In an optional mode, when the recommendation answer is obtained, a video or an image matched with the recommendation question and the recommendation answer can be obtained; further, the video or image may be presented through a live room interface and the virtual anchor lecture recommendation answer is driven.
In addition, as shown in fig. 4B, in the live room interface, optionally, an interaction input area (including but not limited to a comment area) for the audience user to interact with the virtual anchor during the live process can also be displayed; acquiring the interactive input operation of audience users in an interactive input area; and sending the information of the interactive input operation to the server so that the server updates the user information of the corresponding audience user according to the interactive input operation. Therefore, the real-time performance and accuracy of the user information are guaranteed.
Optionally, in this embodiment, the recommendation question is determined according to user information of the audience user, a current live object of the virtual main broadcast explanation, and a question-answer mapping relationship, where the question-answer mapping relationship represents a corresponding relationship between a plurality of live objects and a plurality of question-answer pairs, or the question-answer mapping relationship represents a corresponding relationship between a plurality of live objects and a plurality of question-answer pairs and a plurality of question-answer intentions.
Through the embodiment, the whole question asking process is simplified by recommending the possible question asking recommendation questions for the audience users, the audience users can answer the questions in the live broadcast room only by clicking the corresponding recommendation questions, the question asking efficiency of the audience users and the interaction efficiency between the audience users and the virtual main broadcast are greatly improved, and the question asking cost of the audience users is reduced. When the question answering reply is carried out by combining the video or the image, the experience of audience users in watching the communication in a live broadcast room can be enhanced. It should be understood that the present embodiments may be applicable to various types or scenarios of live rooms, including but not limited to e-commerce, education, culture, sports, news, knowledge, life, electronic contests, etc., and that live question-answer recommendation and question-answer pair construction may be implemented using the above disclosed embodiments and possible variations.
In addition, the description of some steps in this embodiment is simple, and the relevant portions may refer to the description in the foregoing embodiments.
Example four
Referring to fig. 5, a schematic structural diagram of an electronic device according to a fifth embodiment of the present application is shown, and the specific embodiment of the present application does not limit a specific implementation of the electronic device.
As shown in fig. 5, the electronic device may include: a processor (processor)502, a Communications Interface 504, a memory 506, and a communication bus 508.
The processor 502, communication interface 504, and memory 506 communicate with one another via a communication bus 508.
The communication interface 504 is used for communication with other electronic devices or servers.
The processor 502 is configured to execute the program 510, and may specifically execute the relevant steps of any of the above-described method embodiments.
In particular, program 510 may include program code that includes computer operating instructions.
The processor 502 may be a CPU, or an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement embodiments of the present application. The intelligent device comprises one or more processors which can be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.
And a memory 506 for storing a program 510. The memory 506 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 510 may be specifically configured to enable the processor 502 to execute operations corresponding to any one of the methods described in the above method embodiments.
For specific implementation of each step in the program 510, reference may be made to corresponding steps and corresponding descriptions in units in the foregoing method embodiments, and corresponding beneficial effects are provided, which are not described herein again. It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described devices and modules may refer to the corresponding process descriptions in the foregoing method embodiments, and are not described herein again.
The embodiment of the present application further provides a computer program product, which includes computer instructions for instructing a computing device to execute an operation corresponding to any one of the methods in the foregoing method embodiments.
It should be noted that, according to the implementation requirement, each component/step described in the embodiment of the present application may be divided into more components/steps, and two or more components/steps or partial operations of the components/steps may also be combined into a new component/step to achieve the purpose of the embodiment of the present application.
The above-described methods according to embodiments of the present application may be implemented in hardware, firmware, or as software or computer code storable in a recording medium such as a CD ROM, a RAM, a floppy disk, a hard disk, or a magneto-optical disk, or as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium downloaded through a network and to be stored in a local recording medium, so that the methods described herein may be stored in such software processes on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware such as an ASIC or FPGA. It will be appreciated that a computer, processor, microprocessor controller, or programmable hardware includes memory components (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code that, when accessed and executed by a computer, processor, or hardware, implements the methods described herein. Further, when a general-purpose computer accesses code for implementing the methods illustrated herein, execution of the code transforms the general-purpose computer into a special-purpose computer for performing the methods illustrated herein.
Those of ordinary skill in the art will appreciate that the various illustrative elements and method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the embodiments of the present application.
The above embodiments are only used for illustrating the embodiments of the present application, and not for limiting the embodiments of the present application, and those skilled in the relevant art can make various changes and modifications without departing from the spirit and scope of the embodiments of the present application, so that all equivalent technical solutions also belong to the scope of the embodiments of the present application, and the scope of patent protection of the embodiments of the present application should be defined by the claims.

Claims (14)

1. A live question-answering method is suitable for a live broadcast room for live broadcast by utilizing a virtual anchor, and comprises the following steps:
acquiring a current live broadcast object of virtual main broadcast explanation in a live broadcast room and user information of a plurality of audience users, wherein the user information comprises attribute information and/or user preference information of the audience users;
determining different recommended questions for the plurality of audience users according to the current live broadcast object, the user information and a question-answer mapping relation, wherein the question-answer mapping relation represents the corresponding relation between a plurality of live broadcast objects and a plurality of question-answer pairs;
displaying different recommendation problems on live broadcast interfaces of different audience users to recommend the different recommendation problems to the corresponding audience users;
and displaying the corresponding recommended answer on the live broadcast interface of the audience user according to the feedback input of the audience user.
2. The method of claim 1, wherein the determining different recommendation questions for the plurality of viewer users according to the current live object, the user information, and a question-answer mapping relationship comprises:
for each of the viewer users:
analyzing intention of the audience user according to the current live broadcast object and the user information to obtain corresponding user intention;
obtaining a plurality of question-answer pairs corresponding to the current live broadcast object according to the question-answer mapping relation;
acquiring a plurality of question-answer intention information corresponding to the question-answer pairs;
and determining the recommendation questions corresponding to the audience users according to the matching degrees of the user intentions and the question-answer intention information.
3. The method of claim 2, wherein the method further comprises:
determining a live broadcast object corresponding to each question-answer pair based on the question data and the answer data of the question-answer pairs to establish the question-answer mapping relation; and/or
And confirming question-answer intention information corresponding to each question-answer pair based on question data and answer data in the question-answer pairs.
4. The method of claim 1, wherein the presenting the corresponding recommended answer at the live interface of the audience user according to the feedback input of the audience user comprises:
displaying the recommendation answer on a live broadcasting interface of the audience user; and/or
And driving the virtual main broadcasting to explain the recommended answer.
5. The method of claim 4, wherein,
according to the feedback input of the audience user, displaying the corresponding recommendation answer on the live broadcasting interface of the audience user, and further comprising: determining a matching video or image based on the recommended question and the recommended answer; displaying the video or image at a live room interface of the audience user;
wherein the determining a matching video or image based on the recommended question and the recommended answer comprises: and determining a matched video or image based on the semantic labels respectively corresponding to the recommended question and the recommended answer.
6. The method according to claim 1 or 3, wherein the question-answer mapping relation is generated in advance by:
acquiring historical live broadcast data, wherein the historical live broadcast data at least comprises live broadcast audio data and text data corresponding to the live broadcast audio data;
performing question and answer recognition on the text data to obtain question and answer pairs in the text data;
analyzing the question-answer pair to acquire the information of a corresponding live broadcast object;
and establishing a mapping relation between the live broadcast object and the question-answer pair.
7. The method of claim 6, wherein the performing question-answer recognition on the text data to obtain question-answer pairs in the text data comprises:
performing question-answer recognition on the text data to acquire all candidate question-answer pairs in the text data;
performing semantic recognition on all candidate question-answer pairs;
and screening effective question-answer pairs from all candidate question-answer pairs according to the result of semantic recognition.
8. The method of claim 7, wherein the performing question-answer recognition on the text data to obtain question-answer pairs in the text data further comprises:
obtaining effective answer data from the effective question-answer pair;
regenerating updated question data matched with the effective answer data based on the effective answer data;
and taking the updated question data as the question data of the question-answer pair.
9. The method of claim 7, wherein the performing question-answer recognition on the text data to obtain question-answer pairs in the text data further comprises:
determining the relevance of effective answer data in the effective question-answer pair and at least one key sentence;
screening key answer short sentences according to the relevance;
rewriting effective answer data in the effective question-answer pairs according to the key answer short sentences to generate updated answer data, wherein the updated answer data comprises the key answer short sentences;
and using the updated answer data as answer data of the question-answer pair.
10. A live broadcast room interface display method is suitable for a live broadcast room for live broadcast by utilizing a virtual anchor, and comprises the following steps:
after detecting that a plurality of audience users enter a live broadcast room, displaying a live broadcast room interface to the plurality of audience users;
displaying a live broadcasting object currently explained by a virtual main broadcasting of a live broadcasting room in a first area of a live broadcasting room interface of different audience users, and displaying different recommendation problems aiming at the live broadcasting object aiming at different audience users in a second area of the live broadcasting room interface;
the recommendation question is determined according to the user information of the audience users, the current live broadcast object of the virtual main broadcast explanation and a question-answer mapping relation, wherein the question-answer mapping relation represents the corresponding relation between a plurality of live broadcast objects and a plurality of question-answer pairs.
11. The method of claim 10, wherein the method further comprises:
receiving feedback input of the audience user for the presented recommendation question;
acquiring a recommendation answer corresponding to the recommendation question according to the feedback input;
and explaining the recommended answer through the virtual main broadcasting and/or displaying the recommended answer through the live broadcasting room interface.
12. The method of claim 11, wherein,
the obtaining of the recommended answer corresponding to the recommended question according to the feedback input further includes: acquiring a video or an image matched with the recommended question and the recommended answer;
the explaining the recommended answer through the virtual main broadcasting and/or displaying the recommended answer through the live broadcasting room interface further includes: and displaying the video or the image through the live broadcast room interface, and driving the virtual main broadcast to explain the recommended answer.
13. The method of claim 10, wherein the method further comprises:
displaying an interactive input area for the audience user to interact with the virtual anchor in a live broadcasting process on the live broadcasting room interface;
acquiring the interactive input operation of the audience user in the interactive input area;
and sending the information of the interactive input operation to a server so that the server updates the user information of the corresponding audience user according to the interactive input operation.
14. A computer storage medium having stored thereon a computer program which, when executed by a processor, carries out the method of any one of claims 1 to 13.
CN202210067927.1A 2022-01-20 2022-01-20 Live question and answer and interface display method and computer storage medium Pending CN114430490A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210067927.1A CN114430490A (en) 2022-01-20 2022-01-20 Live question and answer and interface display method and computer storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210067927.1A CN114430490A (en) 2022-01-20 2022-01-20 Live question and answer and interface display method and computer storage medium

Publications (1)

Publication Number Publication Date
CN114430490A true CN114430490A (en) 2022-05-03

Family

ID=81314020

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210067927.1A Pending CN114430490A (en) 2022-01-20 2022-01-20 Live question and answer and interface display method and computer storage medium

Country Status (1)

Country Link
CN (1) CN114430490A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114866818A (en) * 2022-06-17 2022-08-05 深圳壹账通智能科技有限公司 Video recommendation method and device, computer equipment and storage medium
CN115190336A (en) * 2022-06-27 2022-10-14 青软创新科技集团股份有限公司 Crowdsourcing type network teaching oriented live broadcast connection method and system
CN115412745A (en) * 2022-08-12 2022-11-29 联想(北京)有限公司 Information processing method and electronic equipment
WO2023221818A1 (en) * 2022-05-17 2023-11-23 北京有竹居网络技术有限公司 Information display method and apparatus, and device and medium
CN114866818B (en) * 2022-06-17 2024-04-26 深圳壹账通智能科技有限公司 Video recommendation method, device, computer equipment and storage medium

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102722558A (en) * 2012-05-29 2012-10-10 百度在线网络技术（北京）有限公司 User question recommending method and device
JP2014112316A (en) * 2012-12-05 2014-06-19 Kddi Corp Question answering program using large amounts of comment sentences, server and method
US20140358720A1 (en) * 2013-05-31 2014-12-04 Yahoo! Inc. Method and apparatus to build flowcharts for e-shopping recommendations
CN106327298A (en) * 2016-08-18 2017-01-11 北京奇虎科技有限公司 Method and device for providing virtual gift recommendation information
CN108536852A (en) * 2018-04-16 2018-09-14 上海智臻智能网络科技股份有限公司 Question and answer exchange method and device, computer equipment and computer readable storage medium
US20180278553A1 (en) * 2015-09-01 2018-09-27 Samsung Electronics Co., Ltd. Answer message recommendation method and device therefor
CN110072116A (en) * 2019-05-06 2019-07-30 广州虎牙信息科技有限公司 Virtual newscaster's recommended method, device and direct broadcast server
CN110099283A (en) * 2019-05-09 2019-08-06 广州虎牙信息科技有限公司 Information-pushing method, device, equipment and storage medium
CN112070524A (en) * 2020-07-24 2020-12-11 广州阿凡提电子科技有限公司 Advertisement service recommendation method and device
CN112148939A (en) * 2019-06-28 2020-12-29 阿里巴巴集团控股有限公司 Data processing method and device and electronic equipment
CN113421143A (en) * 2020-10-10 2021-09-21 阿里巴巴集团控股有限公司 Processing method and device for assisting live broadcast and electronic equipment
CN113641937A (en) * 2021-08-17 2021-11-12 杭州时趣信息技术有限公司 Comment automatic reply method, system and storage medium
CN113747249A (en) * 2021-07-30 2021-12-03 北京达佳互联信息技术有限公司 Live broadcast problem processing method and device and electronic equipment

Patent Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102722558A (en) * 2012-05-29 2012-10-10 百度在线网络技术（北京）有限公司 User question recommending method and device
JP2014112316A (en) * 2012-12-05 2014-06-19 Kddi Corp Question answering program using large amounts of comment sentences, server and method
US20140358720A1 (en) * 2013-05-31 2014-12-04 Yahoo! Inc. Method and apparatus to build flowcharts for e-shopping recommendations
US20180278553A1 (en) * 2015-09-01 2018-09-27 Samsung Electronics Co., Ltd. Answer message recommendation method and device therefor
CN106327298A (en) * 2016-08-18 2017-01-11 北京奇虎科技有限公司 Method and device for providing virtual gift recommendation information
WO2019201098A1 (en) * 2018-04-16 2019-10-24 上海智臻智能网络科技股份有限公司 Question and answer interactive method and apparatus, computer device and computer readable storage medium
CN108536852A (en) * 2018-04-16 2018-09-14 上海智臻智能网络科技股份有限公司 Question and answer exchange method and device, computer equipment and computer readable storage medium
CN110072116A (en) * 2019-05-06 2019-07-30 广州虎牙信息科技有限公司 Virtual newscaster's recommended method, device and direct broadcast server
CN110099283A (en) * 2019-05-09 2019-08-06 广州虎牙信息科技有限公司 Information-pushing method, device, equipment and storage medium
CN112148939A (en) * 2019-06-28 2020-12-29 阿里巴巴集团控股有限公司 Data processing method and device and electronic equipment
CN112070524A (en) * 2020-07-24 2020-12-11 广州阿凡提电子科技有限公司 Advertisement service recommendation method and device
CN113421143A (en) * 2020-10-10 2021-09-21 阿里巴巴集团控股有限公司 Processing method and device for assisting live broadcast and electronic equipment
CN113747249A (en) * 2021-07-30 2021-12-03 北京达佳互联信息技术有限公司 Live broadcast problem processing method and device and electronic equipment
CN113641937A (en) * 2021-08-17 2021-11-12 杭州时趣信息技术有限公司 Comment automatic reply method, system and storage medium

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023221818A1 (en) * 2022-05-17 2023-11-23 北京有竹居网络技术有限公司 Information display method and apparatus, and device and medium
CN114866818A (en) * 2022-06-17 2022-08-05 深圳壹账通智能科技有限公司 Video recommendation method and device, computer equipment and storage medium
CN114866818B (en) * 2022-06-17 2024-04-26 深圳壹账通智能科技有限公司 Video recommendation method, device, computer equipment and storage medium
CN115190336A (en) * 2022-06-27 2022-10-14 青软创新科技集团股份有限公司 Crowdsourcing type network teaching oriented live broadcast connection method and system
CN115190336B (en) * 2022-06-27 2023-02-28 青软创新科技集团股份有限公司 Crowdsourcing type network teaching oriented live broadcast connection method and system
CN115412745A (en) * 2022-08-12 2022-11-29 联想(北京)有限公司 Information processing method and electronic equipment
CN115412745B (en) * 2022-08-12 2024-02-27 联想(北京)有限公司 Information processing method and electronic equipment

Similar Documents

Publication Publication Date Title
CN114430490A (en) Live question and answer and interface display method and computer storage medium
US20050223315A1 (en) Information sharing device and information sharing method
US20030191682A1 (en) Positioning system for perception management
CN112073741A (en) Live broadcast information processing method and device, electronic equipment and storage medium
CN114095749B (en) Recommendation and live interface display method, computer storage medium and program product
US20120135389A1 (en) Learning environment with user defined content
CN113824972B (en) Live video processing method, device, equipment and computer readable storage medium
US20160117301A1 (en) Annotation sharing system and method
CN114339285B (en) Knowledge point processing method, video processing method, device and electronic equipment
CN107071554B (en) Method for recognizing semantics and device
CN111279404B (en) Language fluent system
CN111432282B (en) Video recommendation method and device
US20170171621A1 (en) Method and Electronic Device for Information Processing
US10776415B2 (en) System and method for visualizing and recommending media content based on sequential context
CN107135253A (en) A kind of live course answer method, system and server
CN113254708A (en) Video searching method and device, computer equipment and storage medium
CN112163560A (en) Video information processing method and device, electronic equipment and storage medium
CN111739358A (en) Teaching file output method and device and electronic equipment
JP2009116519A (en) Personal history development device
CN114707502A (en) Virtual space processing method and device, electronic equipment and computer storage medium
CN112732951A (en) Man-machine interaction method and device
US9547995B1 (en) Dynamic instructional course
US20150213726A1 (en) System and methods for automatic composition of tutorial video streams
CN114390306A (en) Live broadcast interactive abstract generation method and device
CN114065057A (en) Search result determining method, display method, device, equipment and medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination